Archive for Personal Finance

Efficient Market Hypothesis in India

Efficient market hypothesis says that stock markets are informationally efficient i.e. all publicly available information is already reflected in the stock prices. (Note that it is mostly illegal to use non-public information for getting advantage on the stock market.) Hence, no strategy can significantly beat the returns of the market (represented by indices) for a sustained time period. People, both professional and lay investor alike, make mammoth efforts in beating the markets, losing sleep over every movement of the market. The efficient market hypothesis, if true, would prove that all such effort is wasted and can at best equal the returns from passive index investing and at worst earn less returns than even the indices.

Studies to prove/disprove this hypothesis have mostly taken place in developed markets. Here, we analyze whether it holds true in Indian markets. Among the different types of people who try and beat the returns of stock market indices, a type that can be easily studied is mutual fund managers. This is because they have to necessarily provide information about their portfolios and their periodic returns due to SEBI guidelines. Mutual fund managers must beat the market indices to justify their high salaries and the expense ratios of mutual funds. It is believed that less than 15% , or 3%! of the mutual funds in the US beat the market indices, which is commonly used as an argument in favour of the efficient market hypothesis. Let us see how Indian mutual funds fare:

I will compare the long term returns of ALL equity diversified mutual funds and ELSS funds with a well known index. Funds for which such analysis is not possible will be discarded and reasons for doing so will be one of the following:

  1. Bonus issues by mutual funds: Date after last bonus issue will be considered. Bonus issues have gone out of fashion these days, so for most funds considered we can get such bonus free time periods of around a decade or so.
  2. Date of launch of the mutual fund: Only funds launched in the previous millennium (31 December 1999 or before) are considered so that we do not get dazzled by short term brilliance.
  3. Date of introduction of the index: In India, our indices are so young that we have mutual funds that are elder than Nifty by a decade or so. Still we can compare the performance of mutual funds against indices for a period of about a decade, so we will go ahead and compare with an existing index.
  4. Total returns: In regular indices, the dividends paid by the constituent companies of the index are not counted in returns. We have considered Nifty as the index for our purposes because its total returns index is readily available, which takes into account the dividend income as well. Since this data is only available 30 June 1999 onwards, mutual fund data of before this date has been discarded.
  5. Sector / Theme specific mutual funds will be discarded.

Following tables summarize our findings. The funds which have beaten Nifty are shown in green, those which lagged Nifty are shown in red. Data comes from Mutual Funds India. Returns have been considered starting from “Start date”, up to 15 July 2009 for the mutual fund as well as the Nifty total returns index.

Diversified Equity Schemes(all Growth/Cumulative option):

Scheme Name Start date Return from scheme (%) Returns from Nifty (Total Returns Index) %
BSL Equity 30/06/1999 1178.83 416.78
Tata Pure Equity 30/06/2000 472.47 332.77
Templeton India Growth 02/05/2000 651.73 367.63
HDFC Equity 30/06/1999 1316.3 416.78
Franklin India Prima Plus 30/06/1999 1070.69 416.78
Taurus Starshare 30/06/1999 598.16 416.78
Franklin India Bluechip 24/03/2000 523.58 314.09
HDFC Top 200 27/03/2000 518.97 315.41
Reliance Growth 30/06/1999 1766.29 416.78
LIC MF Equity 30/06/1999 225.91 416.78
Franklin India Prima 30/06/1999 1143.69 416.78
HDFC Capital Builder 30/06/1999 584.77 416.78
Tata Growth 14/07/1999 718.47 370.52
LIC MF Growth 30/06/1999 247.52 416.78
Taurus Discovery 14/07/1999 249.45 370.52
JM Equity 30/06/1999 283.31 416.78
BSL Advantage 23/03/2000 168.94 317.35
Reliance Vision 30/06/1999 1350.91 416.78
ICICI Prudential Growth 30/06/1999 695.17 416.78
ING Core Equity 30/06/1999 272.1 416.78
BSL MNC 28/12/1999 312.36 343.92
Sundaram BNP Paribas Growth Reg 16/05/2000 496.67 375.03

Equity Linked Savings Schemes(all Growth/Cumulative option):

Scheme Name Start date Return from scheme (%) Returns from Nifty (Total Returns Index) %

HDFC Taxsaver

06/04/00

747.82

338.08

LIC MF Tax Plan

30/06/1999

209.75

416.78

Franklin India Tax Shield

30/06/1999

1319.07

416.78

ICICI Prudential Taxplan

19/08/1999

852.5

367.65

These findings deal a fatal blow to the efficient market theory. Not only do about 70% of the mutual funds beat the market, many of them do so by a factor of 2 or more. Reliance Growth beat the market in this period by more than 400%. Note that the index returns considered are directly from the index and not from any index fund. If an index fund is used for investment, there will be some expense ratio of the fund and hence the index returns will be even lower than mentioned above.

This article is not intended to advertize mutual funds. Nor am I saying that mutual funds will necessarily continue beating the market by a wide margin. I have only analyzed the available data and come to the conclusion that Indian markets have not been demonstrably informationally efficient over the last decade. Hence stock picking has a lot of value in India. All the lost sleep over your stocks was not in vain, after all :).

Limitations of the above comparison:

  1. The comparison suffers from a survivorship bias. Which is to say, the funds which were performing very poorly were discontinued long ago and hence they do not appear in this study. Though in my defence, we have not seen a huge number of mutual fund schemes being discontinued, so possibly the bias does not invalidate my conclusions. In the US, only 15% or 3% funds beat the index even after a survivorship bias.
  2. A decade is not enough to get meaningful conclusions about long term trends. Here my defence is stronger. Efficient market hypothesis categorically denies the possibility of any strategy being capable of beating the market. As such, the onus is on the propounders of the hypothesis to prove that it is so. As market indices in India are very young, supporters of efficient market hypothesis also do not have enough data to be able to prove the hypothesis beyond reasonable doubt. All I claim is, the short term data that we have does not support the efficient market hypothesis.
  3. Maybe, NOW the markets are efficient and going forward it might get difficult to beat the market indices. Agreed. We can only say that so far there is no support for efficient market hypothesis in India. If you are reading this in year 2100 or more, maybe you can throw some light on this matter in the comments section.

Other problems with index funds in India:

  1. High tracking errors: Tracking errors as high as 10% are not unheard of, and fund managers have the audacity to justify the tracking errors. This proves that the index funds were started just to jump on the bandwagon but in essence many of the index funds are being actively managed.
  2. High expense ratios: Most index funds charge an expense ratio of about 1 – 1.5% which is outrageous. The exception is in ETFs (exchange traded funds) which charge around 0.5%. But even this is too high as there is not much that the fund manager has to do with an index fund and a lot of it can be automated by the fund house. Main reason to invest in an index fund is the low costs. In the developed markets, less than 0.25% expense ratio is charged for index funds. Hopefully as more money gets invested into index funds, fund houses will get a better economy of scale and expense ratios will come down.
    Chicken and egg: Index funds in India are not a good idea, partly because of high expense ratios, and expense ratios are high partly because not many invest in index funds because index funds are not a good idea.

***************

Appendix

More support about passive investing from developed markets:

http://www.stanford.edu/~wfsharpe/art/talks/indexed_investing.htm

http://seekingalpha.com/article/84916-percentage-of-active-fund-managers-that-can-beat-the-market-shrinking-rapidly

http://allfinancialmatters.com/2008/01/04/when-an-active-mutual-fund-manager-beats-the-market-is-it-luck/

Comments (3)

NPS: Not cheap

PFRDA (Pension Fund Regulatory and Development Authority) has opened the NPS (New Pension Scheme) for general public. Note that it has been “New” for last 2 decades and has materialized only now.

The idea of NPS is extremely noble: India lacks social security of any kind, so the unorganized sector badly needs a useful retirement savings scheme. But the current implementation leaves a lot to be desired. Only saving grace is, that there is no one to hard sell the scheme yet. The appointed intermediaries have no interest in selling the scheme, instead, they have competing products of their own to sell. Since not many people will open an NPS account until some sellers are motivated, not many people will be disappointed that NPS is not cheap.

The promises of fund management charges of 0.009% sounds very cheap. Much cheaper than mutual funds which can charge up to 2.5% fund management charges per year (most charge less than 2%). But there are other problems that we now proceed to analyze.

Target Audience:

Our first step is to understand the target beneficiaries of NPS. Most organized sector workers of lower-middle to middle class (income higher than 1.5 lakhs a year) don’t need NPS that badly because they already have EPF,  government pension(government employees who joined service a decade ago) and other loyalty funds from their employer such as superannuation. Though they can still use NPS to get some equity exposure as their other investments are likely to be largely fixed income investments giving lower returns (less than 9% post tax).

The prime beneficiaries for whom NPS could have been a godsend are unorganized sector workers who do not have access to any retirement scheme. They are also typically poorer than organized sector workers, and people with incomes 75000 – 1,50,000 could have taken advantage of this scheme (if it were sold well, but that is a separate topic). Let us concentrate on such people for our analysis. Yes, I am less concerned with “high” income people, earning more than 1.5 lakhs a year as they can better fend for themselves.

Why I am leaving out people earning less than 75000 per year? Depending on circumstances (rural/urban ; family situation etc.) they may not have any invesible surplus. Even for such important a task as retirement planning. So they could not have taken advantage of NPS even if NPS were perfect.

I know there are exceptions to the generalities I quoted above. There are people in unorganized sector that earn crores a year. There are people earning 50,000 a year who can afford to save because of low expenses. But, like I said, they are exceptions rather than the rule and hence can be ignored for the larger analysis.
Disadvantages of NPS for our specific target audience:

  1. Compulsary investment of 6000 a year:
    A commitment of investing Rs. 6000 every year is tough to make. More so because our target audience has unstable sources of income. Rs. 6000 is anywhere from 8% to 4% of the annual income of our target audience and hence they might be very scared to make such commitments. Especially since the penalty of Rs. 100 per quarter if this minimum investment is not made is very high. On the flip side, it can be argued that this penalty keeps them disciplined; though I wouldn’t buy this argument for much. Retirement planning takes mammoth discipline anyway. Since it is important that our target audience starts some investment right away to take advantage of the power of compounding, to start with we should make our schemes very flexible, and low on penalties.
  2. Charges (9.33% entry load):
    Fund management charges are low enough, but the fixed charges are high. In a bad year, when you barely manage to invest 6000 in the requisite 4 yearly installments, you incur the following charges:
    A. Account opening charge of Rs. 50 (only required for the first year)
    B. Annual maintenance charge of Rs. 350
    C. 4 transactions, Rs. 10 fees to CRA for each. Total Rs. 40.
    D. Registration with PoP (Point of Presence, kind of the investor’s broker): Rs. 40. This will be required not just the first time, but everytime the PoP is changed for some reason (e.g. migration from one location to another)
    D. 4 transactions, Rs. 20 fees to PoP for each. Total Rs. 80.
    Other charges are negligible; but these charges total to Rs. 560. This is 9.33 percentage of the investment for the year (Rs. 6000). Consider it kind of an entry load for NPS.
  3. Expensive modes of withdrawal:
    The funds are reasonably high return : 50% stock market index funds & 50% debt, progressively moving towards majority debt as the individual grows older. But remember that taxes will kill the returns. Unless an annuity is purchased from a life insurance company. Annuity rates given by LIC are on the lower side (those by ICICI Prudential are even lower).

In conclusion, we can say that the high returns from stock markets will be eroded because of high charges and annuity inefficiencies. This article is not to totally condemn the designers of NPS but just to show that more needs to be done. The charges will be reduced only when there are millions of investors in NPS; which will be never if these defects are not rectified. Which brings us to our original point: there is no one to explain all these nitty-gritties to an investor in our target audience. No significant people of our target audience is going to invest in NPS anyway, so all the above rant was for academic purposes only. Government employees, and organized sector workers’ EPF will be replaced with NPS which is good thing.

Comments (3)

I don’t need Life insurance

Everybody and his dog will tell you that you need life insurance. I won’t even go into ULIPs as they are clearly devil incarnate. Even pure term insurance is not what everybody needs. I will summarize below the “reasons” people give for recommending a life insurance policy to you. Even though you don’t have any financial dependants. After a while, it becomes so irritating it is not even funny. The last “reason” is the best: even seasoned, intelligent, honest personal finance writers would tell you this one.

1. Tax benefit :

Are you for real? You get tax benefit on the money that you don’t even receive. It is like quitting your job to save taxes. Allow me to invent a sardar joke:

I once met a sardarji who was standing near a well and tearing and throwing 100 rupee notes in the well. I went to him and asked, “Sardarji, why are you throwing money down the well?”. The inimitable sardarji replied: “Bhai saab, for every 100 rupee note that I throw down the well, the government gives me 30 rupees tax break”.

2. You are worth it:

Yep, I am worth it. That is why I am spending it on a vacation in Switzerland. Or taking my girlfriend (earning well, NOT dependent on me in the least bit) to dinners in  the Zodiac Grill, Taj Mahal Palace. NOT buying a stupid life insurance policy.

3. Premiums are lower if you start young:

Most unsuspecting suckers would fall for this. The beauty of this is: it sounds correct. And the real calculations are a bit complex. Not complex enough that you can’t do them but just enough that you get lazy and get suckered into buying a policy. If you don’t want to go into calculations, here is the crux of the counter argument: the premiums are ZERO when you are not insured. This is lower than any life insurance premium you will come across. These saved premiums can even be invested to earn you more money.

Since the calculations are “complex” you are too lazy to do them. But not to worry, I don’t call myself Business Pandit for no reason. Lets get cracking:

The example considered here is from SBI Life insurance. (The website of LIC is down today). There is this Mr. Sucker, 25 year old in 2009. No dependants. Question is to insure Mr. Sucker for Rs. 10 lakhs up to an age of 40 years(2024). He acquires a dependant (say, Mrs Sucker, and no, it’s not funny) when he is 30 years old(2014). Competition is between the following 2 ideas:

A. Get a life insurance at an age of 25 to take advantage of the “low” premiums. This is a policy for 15 years. Premium is Rs. 1954 per annum. Total premium Rs. 29,310.

B. Remain uninsured until the age of 30, and then get a policy for 10 years. Premium is ZERO for the first 5 years and Rs. 2150 for the next 10 years. Total premium 21,500.

Money saved with plan B is Rs. 7,810. This is not even considering that most of the money was saved in the earlier part of the plan which could have been invested. Skip the following block if calculations bore you.

Let us calculate the total cost of insuring Mr. Sucker in year 2024 rupees. Post-tax pre-inflation rate of return considered is 8%.
Cost of plan A = 1954*(1.08^14) + 1954*(1.08^13) + … + 1954

= 1954*(1.08^15 – 1)/(1.08-1)

= 53055

Cost of plan B = 2150*(1.08^9) + 2150*(1.08^8) + … + 2150

= 2150*(1.08^10 – 1)/(1.08-1)

= 31146

So in 2024 value of rupee, 53055-31146 = 21909 rupees are saved.

Philosophy:

1. Traditional media will not tell you the dangers of life insurance mis-selling. Life insurance companies are big advertisement spenders. Pissing them off is dangerous for any advertisement funded media (newspaper / magazine / TV channel / websites that write their own content). It is only some bloggers who will tell you the truth as we are not dependent on advertisement for our livelihood.

2. Yes India is severely under-insured. Yes we as a society need to encourage life insurance companies. No, this does not mean allowing 420 behaviour in life insurance.

Comments (6)